A novel replica placement strategy using binary item-to-item collaborative filtering for efficient voronoi-based cloud-oriented content delivery network

Abstract

On-demand cloud-oriented content delivery service provision is the challenging task for the most of researchers. Antecedent end users used to request to the central server for accessing information but now a day's massive use of mobile devices upshots delivery of intended contents virtually from edge servers which are scattered over the earth surface. Thus, placing content effectively to appropriate edge servers and searching it from edge servers are two big issues for cloud-oriented content delivery network providers (CCDNPs). To solve these issues, at first we have clustered all edge servers using voronoi diagram to locate each edge server into a particular voronoi cell and also search the requested content from nearest edge server of the requested user current location using Delaunay triangulation over our aforementioned clustering technique. Consequently we have also used binary item-to-item collaborative filtering for placing content based upon the users' interest. Our proposed content searching technique and replica placement strategy not only solve the users' requirement but also overcome the open challenges of the CCDNPs.